From: Cliff White <cliffw@osdl.org>
To: "Martin J. Bligh" <mbligh@aracnet.com>
Cc: Chris Wedgwood <cw@f00f.org>, Alan Cox <alan@lxorguk.ukuu.org.uk>,
Larry McVoy <lm@bitmover.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
cliffw@osdl.org
Subject: Re: Minutes from Feb 21 LSE Call
Date: Tue, 25 Feb 2003 09:11:38 -0800 [thread overview]
Message-ID: <200302251711.h1PHBct16624@mail.osdl.org> (raw)
In-Reply-To: Message from "Martin J. Bligh" <mbligh@aracnet.com> of "Mon, 24 Feb 2003 22:17:05 PST." <13760000.1046153824@[10.10.2.4]>
> >> _If_ it harms performance on small boxes.
> >
> > You mean like the general slowdown from 2.4 - >2.5?
> >
> > It seems to me for small boxes, 2.5.x is margianlly slower at most
> > things than 2.4.x.
>
> Can you name a benchmark, or at least do something reproducible between
> versions, and produce a 2.4 vs 2.5 profile? Let's at least try to fix it ...
>
> M.
Well, here's one bit of data. Easy enough to do if you have a web browser.
LMBench 2.0 on 1-way and 2-way, kernels 2.4.18 and 2.5.60
1-way (stp1-003 stp1-002)
2.4.18 http://khack.osdl.org/stp/7443/
2.5.60 http://khack.osdl.org/stp/265622/
2-way (stp2-003 stp2-000)
2.4.18 http://khack.osdl.org/stp/3165/
2.5.60 http://khack.osdl.org/stp/265643/
Interesting items for me are the fork/exec/sh times and some of the file + VM
numbers
LMBench 2.0 Data ( items selected from total of five runs )
Processor, Processes - times in microseconds - smaller is better
----------------------------------------------------------------
Host OS Mhz null null open selct sig sig fork exec sh
call I/O stat clos TCP inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ----- ---- ---- ---- ---- ----
stp2-003. Linux 2.4.18 1000 0.39 0.67 3.89 4.99 30.4 0.93 3.06 344. 1403 4465
stp2-000. Linux 2.5.60 1000 0.41 0.77 4.34 5.57 32.6 1.15 3.59 245. 1406 5795
stp1-003. Linux 2.4.18 1000 0.32 0.46 2.60 3.21 16.6 0.79 2.52 104. 918. 4460
stp1-002. Linux 2.5.60 1000 0.33 0.47 2.83 3.47 16.0 0.94 2.70 143. 1212 5292
Context switching - times in microseconds - smaller is better
-------------------------------------------------------------
Host OS 2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
--------- ------------- ----- ------ ------ ------ ------ ------- -------
stp2-003. Linux 2.4.18 2.680 6.2100 15.8 7.9400 110.7 26.4 111.1
stp2-000. Linux 2.5.60 1.590 5.0700 17.6 7.5800 79.8 11.0 113.6
stp1-003. Linux 2.4.18 0.590 3.4700 11.1 4.8200 134.3 30.8 131.7
stp1-002. Linux 2.5.60 1.000 3.5400 11.2 4.1400 129.6 30.4 127.8
*Local* Communication latencies in microseconds - smaller is better
-------------------------------------------------------------------
Host OS 2p/0K Pipe AF UDP RPC/ TCP RPC/ TCP
ctxsw UNIX UDP TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
stp2-003. Linux 2.4.18 2.680 9.071 17.5 26.9 46.2 34.4 60.0 62.9
stp2-000. Linux 2.5.60 1.590 8.414 13.2 21.2 43.2 28.3 54.1 97.1
stp1-003. Linux 2.4.18 0.590 3.623 6.98 11.7 28.2 17.8 38.4 300K
stp1-002. Linux 2.5.60 1.050 4.591 8.54 14.8 31.8 20.0 41.0 67.1
File & VM system latencies in microseconds - smaller is better
--------------------------------------------------------------
Host OS 0K File 10K File Mmap Prot Page
Create Delete Create Delete Latency Fault Fault
--------- ------------- ------ ------ ------ ------ ------- ----- -----
stp2-003. Linux 2.4.18 34.6 7.2490 110.9 17.9 2642.0 0.771 3.00000
stp2-000. Linux 2.5.60 40.0 9.2780 113.3 23.3 4592.0 0.543 3.00000
stp1-003. Linux 2.4.18 28.8 4.8890 107.5 11.3 686.0 0.621 2.00000
stp1-002. Linux 2.5.60 32.4 6.4290 112.9 16.2 1455.0 0.465 2.00000
*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------
Host OS Pipe AF TCP File Mmap Bcopy Bcopy Mem Mem
UNIX reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
stp2-003. Linux 2.4.18 563. 277. 263. 437.0 552.8 249.1 180.7 553. 215.2
stp2-000. Linux 2.5.60 603. 516. 151. 436.3 549.0 238.0 171.9 548. 233.7
stp1-003. Linux 2.4.18 1009 820. 404. 414.3 467.0 167.2 154.1 466. 236.2
stp1-002. Linux 2.5.60 806. 584. 69.1 408.0 461.7 161.1 149.1 461. 233.5
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
---------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Guesses
--------- ------------- ---- ----- ------ -------- -------
stp2-003. Linux 2.4.18 1000 3.464 8.0820 110.9
stp2-000. Linux 2.5.60 1000 3.545 8.2790 110.6
stp1-003. Linux 2.4.18 1000 2.994 6.9850 121.4
stp1-002. Linux 2.5.60 1000 3.023 7.0530 122.5
------------------
cliffw
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
next prev parent reply other threads:[~2003-02-25 17:01 UTC|newest]
Thread overview: 289+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <Pine.LNX.4.44.0302221417120.2686-100000@coffee.psychology.mcmaster.ca>
[not found] ` <1510000.1045942974@[10.10.2.4]>
2003-02-22 19:56 ` Larry McVoy
2003-02-22 20:24 ` William Lee Irwin III
2003-02-22 21:02 ` Martin J. Bligh
2003-02-22 22:06 ` Mark Hahn
2003-02-22 22:17 ` William Lee Irwin III
2003-02-22 23:28 ` Larry McVoy
2003-02-22 23:47 ` Martin J. Bligh
2003-02-23 0:09 ` Gerrit Huizenga
2003-02-23 8:01 ` Larry McVoy
2003-02-23 8:05 ` William Lee Irwin III
2003-02-24 18:36 ` Andy Pfiffer
2003-02-22 22:44 ` Ben Greear
2003-02-23 23:29 ` Bill Davidsen
2003-02-23 23:37 ` Martin J. Bligh
2003-02-24 4:57 ` Larry McVoy
2003-02-24 6:10 ` Gerhard Mack
2003-02-24 6:52 ` Larry McVoy
2003-02-24 7:46 ` Bill Huey
2003-02-24 7:44 ` Bill Huey
2003-02-24 7:54 ` William Lee Irwin III
2003-02-24 8:00 ` Bill Huey
2003-02-24 8:40 ` Andrew Morton
2003-02-24 8:50 ` William Lee Irwin III
2003-02-24 16:17 ` yodaiken
2003-02-24 23:13 ` William Lee Irwin III
2003-02-24 23:27 ` yodaiken
2003-02-24 23:54 ` William Lee Irwin III
2003-02-24 23:54 ` yodaiken
2003-02-25 2:17 ` Bill Huey
2003-02-25 2:24 ` yodaiken
2003-02-25 2:35 ` Bill Huey
2003-02-25 2:43 ` Bill Huey
2003-02-25 2:32 ` Larry McVoy
2003-02-25 2:40 ` Bill Huey
2003-02-25 5:24 ` Rik van Riel
2003-02-25 15:30 ` Alan Cox
2003-02-25 14:59 ` Bill Huey
2003-02-25 15:44 ` yodaiken
2003-02-26 19:31 ` Bill Davidsen
2003-02-27 0:56 ` Bill Huey
2003-02-27 20:04 ` Bill Davidsen
2003-02-25 2:07 ` Bill Huey
2003-02-25 2:14 ` Larry McVoy
2003-02-25 2:24 ` Bill Huey
2003-02-25 2:46 ` Valdis.Kletnieks
2003-02-25 14:47 ` Mr. James W. Laferriere
2003-02-25 15:59 ` Jesse Pollard
2003-02-24 8:56 ` Bill Huey
2003-02-24 9:09 ` Andrew Morton
2003-02-24 9:24 ` Bill Huey
2003-02-24 9:56 ` Andrew Morton
2003-02-24 10:11 ` Bill Huey
2003-02-24 14:40 ` Bill Davidsen
2003-02-24 21:10 ` Andrea Arcangeli
2003-02-24 8:43 ` William Lee Irwin III
2003-02-22 23:10 ` Martin J. Bligh
2003-02-22 23:20 ` Larry McVoy
2003-02-22 23:46 ` Martin J. Bligh
2003-02-25 2:19 ` Hans Reiser
2003-02-25 3:49 ` Martin J. Bligh
2003-02-25 5:12 ` Steven Cole
2003-02-25 20:37 ` Scott Robert Ladd
2003-02-25 21:36 ` Hans Reiser
2003-02-25 23:28 ` Scott Robert Ladd
2003-02-25 23:41 ` Hans Reiser
2003-02-26 0:19 ` Scott Robert Ladd
2003-02-26 0:35 ` Hans Reiser
2003-02-26 16:31 ` Horst von Brand
2003-02-26 0:47 ` Steven Cole
2003-02-26 16:07 ` Horst von Brand
2003-02-26 19:47 ` Alan Cox
2003-02-26 6:04 ` Aaron Lehmann
2003-02-26 0:44 ` Alan Cox
2003-02-25 23:58 ` Scott Robert Ladd
2003-02-22 23:15 ` Larry McVoy
2003-02-22 23:23 ` Christoph Hellwig
2003-02-22 23:54 ` Mark Hahn
2003-02-22 23:44 ` Martin J. Bligh
2003-02-24 4:56 ` Larry McVoy
2003-02-24 5:06 ` William Lee Irwin III
2003-02-24 6:00 ` Mark Hahn
2003-02-24 6:02 ` William Lee Irwin III
2003-02-24 15:06 ` Alan Cox
2003-02-24 23:18 ` William Lee Irwin III
2003-02-24 5:16 ` Martin J. Bligh
2003-02-24 6:58 ` Larry McVoy
2003-02-24 7:39 ` Martin J. Bligh
2003-02-24 16:17 ` Larry McVoy
2003-02-24 16:49 ` Martin J. Bligh
2003-02-25 0:41 ` Server shipments [was Re: Minutes from Feb 21 LSE Call] Larry McVoy
2003-02-25 0:41 ` Martin J. Bligh
2003-02-25 0:54 ` Larry McVoy
2003-02-25 2:00 ` Tupshin Harper
2003-02-25 3:54 ` Martin J. Bligh
2003-02-25 3:00 ` Martin J. Bligh
2003-02-25 3:13 ` Larry McVoy
2003-02-25 4:11 ` Martin J. Bligh
2003-02-25 4:17 ` Larry McVoy
2003-02-25 4:21 ` Martin J. Bligh
2003-02-25 4:37 ` Larry McVoy
2003-02-25 22:02 ` Gerrit Huizenga
2003-02-25 23:19 ` Larry McVoy
2003-02-25 23:46 ` Gerhard Mack
2003-02-26 4:23 ` Jesse Pollard
2003-02-26 5:05 ` William Lee Irwin III
2003-02-26 5:27 ` Bernd Eckenfels
2003-02-26 9:36 ` Eric W. Biederman
2003-02-26 12:09 ` Jesse Pollard
2003-02-26 16:42 ` Geert Uytterhoeven
2003-02-25 17:37 ` Andrea Arcangeli
2003-02-25 1:09 ` David Lang
2003-02-24 18:22 ` Minutes from Feb 21 LSE Call John W. M. Stevens
2003-02-24 7:51 ` William Lee Irwin III
2003-02-24 15:47 ` Larry McVoy
2003-02-24 16:00 ` Martin J. Bligh
2003-02-24 16:23 ` Benjamin LaHaise
2003-02-24 16:25 ` yodaiken
2003-02-24 18:20 ` Gerrit Huizenga
2003-02-25 1:51 ` Minutes from Feb 21 LSE Call - publishing performance data Craig Thomas
2003-02-24 16:31 ` Minutes from Feb 21 LSE Call Larry McVoy
2003-02-24 23:36 ` William Lee Irwin III
2003-02-25 0:23 ` Larry McVoy
2003-02-25 2:37 ` Werner Almesberger
2003-02-25 4:42 ` William Lee Irwin III
2003-02-25 4:54 ` Larry McVoy
2003-02-25 6:00 ` William Lee Irwin III
2003-02-25 7:00 ` Val Henson
2003-02-24 13:28 ` Alan Cox
2003-02-25 5:19 ` Chris Wedgwood
2003-02-25 5:26 ` William Lee Irwin III
2003-02-25 21:21 ` Chris Wedgwood
2003-02-25 21:14 ` Martin J. Bligh
2003-02-25 21:21 ` William Lee Irwin III
2003-02-25 22:08 ` Larry McVoy
2003-02-25 22:10 ` William Lee Irwin III
2003-02-25 22:37 ` Chris Wedgwood
2003-02-25 22:58 ` Larry McVoy
2003-02-25 6:17 ` Martin J. Bligh
2003-02-25 17:11 ` Cliff White [this message]
2003-02-25 17:17 ` William Lee Irwin III
2003-02-25 17:38 ` Linus Torvalds
2003-02-25 19:54 ` Dave Jones
2003-02-26 2:04 ` Linus Torvalds
2003-02-25 19:48 ` Martin J. Bligh
2003-02-25 21:28 ` William Lee Irwin III
2003-02-25 19:20 ` Alan Cox
2003-02-25 19:59 ` Scott Robert Ladd
2003-02-25 20:18 ` jlnance
2003-02-25 20:59 ` Scott Robert Ladd
2003-02-25 21:19 ` Chris Wedgwood
2003-02-25 21:38 ` Scott Robert Ladd
2003-02-24 18:44 ` Davide Libenzi
2003-02-22 23:57 ` Jeff Garzik
2003-02-23 23:57 ` Bill Davidsen
2003-02-24 6:22 ` Val Henson
2003-02-24 6:41 ` William Lee Irwin III
2003-02-22 21:29 ` Jeff Garzik
2003-02-24 2:04 linux
2003-02-24 2:39 ` Linus Torvalds
2003-02-24 3:28 ` David Lang
2003-02-26 5:30 ` Bernd Eckenfels
2003-02-26 5:42 ` William Lee Irwin III
2003-02-26 7:22 ` David Lang
2003-02-27 17:50 ` Daniel Egger
2003-02-27 18:25 ` David Lang
2003-02-28 8:58 ` Filip Van Raemdonck
2003-02-28 19:48 ` Arador
2003-03-01 0:51 ` Chris Wedgwood
2003-03-01 1:14 ` Davide Libenzi
2003-03-01 1:27 ` David Lang
2003-03-01 14:15 ` Daniel Egger
2003-02-24 4:42 ` Martin J. Bligh
2003-02-24 4:58 ` Linus Torvalds
-- strict thread matches above, loose matches on Subject: below --
2003-02-21 23:48 Hanna Linder
2003-02-22 0:16 ` Larry McVoy
2003-02-22 0:25 ` William Lee Irwin III
2003-02-22 2:24 ` Steven Cole
2003-02-22 0:44 ` Martin J. Bligh
2003-02-22 2:47 ` Larry McVoy
2003-02-22 4:32 ` Martin J. Bligh
2003-02-22 5:05 ` Larry McVoy
2003-02-22 6:39 ` Martin J. Bligh
2003-02-22 8:38 ` Jeff Garzik
2003-02-22 22:18 ` William Lee Irwin III
2003-02-23 0:50 ` Martin J. Bligh
2003-02-23 11:22 ` Magnus Danielson
2003-02-23 19:54 ` Eric W. Biederman
2003-02-23 1:17 ` Benjamin LaHaise
2003-02-23 5:21 ` Gerrit Huizenga
2003-02-23 8:07 ` David Lang
2003-02-23 8:20 ` William Lee Irwin III
2003-02-23 19:17 ` Linus Torvalds
2003-02-23 19:29 ` David Mosberger
2003-02-23 20:13 ` Martin J. Bligh
2003-02-23 22:01 ` David Mosberger
2003-02-23 22:12 ` Martin J. Bligh
2003-02-23 21:34 ` Linus Torvalds
2003-02-23 22:40 ` David Mosberger
2003-02-23 22:48 ` David Lang
2003-02-23 22:54 ` David Mosberger
2003-02-23 22:56 ` David Lang
2003-02-24 0:40 ` Linus Torvalds
2003-02-24 2:32 ` David Mosberger
2003-02-24 2:54 ` Linus Torvalds
2003-02-24 3:08 ` David Mosberger
2003-02-24 21:42 ` Andrea Arcangeli
2003-02-24 1:06 ` dean gaudet
2003-02-24 1:56 ` David Mosberger
2003-02-24 2:15 ` dean gaudet
2003-02-24 3:11 ` David Mosberger
2003-02-23 23:06 ` Martin J. Bligh
2003-02-23 23:59 ` David Mosberger
2003-02-24 3:49 ` Gerrit Huizenga
2003-02-24 4:07 ` David Mosberger
2003-02-24 4:34 ` Martin J. Bligh
2003-02-24 5:02 ` Gerrit Huizenga
2003-02-23 20:21 ` Xavier Bestel
2003-02-23 20:50 ` Martin J. Bligh
2003-02-23 23:57 ` Alan Cox
2003-02-24 1:26 ` Kenneth Johansson
2003-02-24 1:53 ` dean gaudet
2003-02-23 21:35 ` Alan Cox
2003-02-23 21:41 ` Linus Torvalds
2003-02-24 0:01 ` Bill Davidsen
2003-02-24 0:36 ` yodaiken
2003-02-23 21:15 ` John Bradford
2003-02-23 21:45 ` Linus Torvalds
2003-02-24 1:25 ` Benjamin LaHaise
2003-02-23 21:55 ` William Lee Irwin III
2003-02-23 19:13 ` David Mosberger
2003-02-23 23:28 ` Benjamin LaHaise
2003-02-26 8:46 ` Eric W. Biederman
2003-02-23 20:48 ` Gerrit Huizenga
2003-02-23 9:37 ` William Lee Irwin III
2003-02-22 8:38 ` David S. Miller
2003-02-22 8:38 ` David S. Miller
2003-02-22 14:34 ` Larry McVoy
2003-02-22 15:47 ` Martin J. Bligh
2003-02-22 16:13 ` Larry McVoy
2003-02-22 16:29 ` Martin J. Bligh
2003-02-22 16:33 ` Larry McVoy
2003-02-22 16:39 ` Martin J. Bligh
2003-02-22 16:59 ` John Bradford
2003-02-24 18:00 ` Timothy D. Witham
2003-02-22 8:32 ` David S. Miller
2003-02-22 18:20 ` Alan Cox
2003-02-22 20:05 ` William Lee Irwin III
2003-02-22 21:35 ` Alan Cox
2003-02-22 21:36 ` Gerrit Huizenga
2003-02-22 21:42 ` Christoph Hellwig
2003-02-23 23:23 ` Bill Davidsen
2003-02-24 3:31 ` Gerrit Huizenga
2003-02-24 4:02 ` Larry McVoy
2003-02-24 4:15 ` Russell Leighton
2003-02-24 5:11 ` William Lee Irwin III
2003-02-24 8:07 ` Christoph Hellwig
2003-02-23 0:37 ` Eric W. Biederman
2003-02-23 0:42 ` Eric W. Biederman
2003-02-23 14:29 ` Rik van Riel
2003-02-23 17:28 ` Eric W. Biederman
2003-02-24 1:42 ` Benjamin LaHaise
2003-02-23 3:24 ` Andrew Morton
2003-02-25 17:17 ` Andrea Arcangeli
2003-02-25 17:43 ` William Lee Irwin III
2003-02-25 17:59 ` Andrea Arcangeli
2003-02-25 18:04 ` William Lee Irwin III
2003-02-25 18:50 ` William Lee Irwin III
2003-02-25 19:18 ` Andrea Arcangeli
2003-02-25 19:27 ` Martin J. Bligh
2003-02-25 20:30 ` Andrea Arcangeli
2003-02-25 20:53 ` Martin J. Bligh
2003-02-25 21:17 ` Andrea Arcangeli
2003-02-25 21:12 ` Martin J. Bligh
2003-02-25 22:16 ` Andrea Arcangeli
2003-02-25 22:17 ` Martin J. Bligh
2003-02-25 22:37 ` Andrea Arcangeli
2003-02-25 21:26 ` William Lee Irwin III
2003-02-25 22:18 ` Andrea Arcangeli
2003-02-26 5:24 ` Rik van Riel
2003-02-26 5:38 ` William Lee Irwin III
2003-02-26 6:01 ` Martin J. Bligh
2003-02-26 6:14 ` William Lee Irwin III
2003-02-26 6:32 ` William Lee Irwin III
2003-02-26 16:02 ` Rik van Riel
2003-02-27 3:48 ` Daniel Phillips
2003-02-25 20:10 ` William Lee Irwin III
2003-02-25 20:23 ` Andrea Arcangeli
2003-02-25 20:46 ` William Lee Irwin III
2003-02-25 20:52 ` Andrea Arcangeli
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200302251711.h1PHBct16624@mail.osdl.org \
--to=cliffw@osdl.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=cw@f00f.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lm@bitmover.com \
--cc=mbligh@aracnet.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®